Jian Wang

Class transfer estimation in neural networks for backdoor target identification and mitigation

Wenbin Jiang, Shikang Xie, Jiqiang Liu, Nan Jiang et autres

cn (code pays fourni par la source)

Mask2Keep: Mask-guided information transfer for backdoors resilient to compression-oriented pruning

Jing Shang, Jian Wang, Kailun Wang, Xu Liu et autres

cn (code pays fourni par la source)

FALCON: Federated Active Learning-Based Concept Drift Adaptation for Malware Detection

Kailun Wang, Ruichen Zhang, Jiqiang Liu, Dusit Tao Niyato et autres

Mobile malware detection faces critical challenges from concept drift, as evolving threat characteristics make static detection models ineffective. Active learning supports continual model adaptation to mitigate concept drift. However, it requires manual annotation. This limits scalability when drift samples grow rapidly. Federated …

cn, sg (code pays fourni par la source)

Ensemble Shapley: Toward an Efficient and Reliable Data Valuation With Guided Ensemble Aggregation

Wenbin Jiang, Jiqiang Liu, Jian Wang, Zhaolin Liu et autres

Data valuation provides a principled framework for quantifying the contribution of data to model training. It plays a crucial role in trustworthy machine learning (ML) by supporting data curation, enhancing interpretability, and enabling fair incentive mechanisms in data markets. Shapley value is …

cn, sg (code pays fourni par la source)

Double-Blind Cleanser: Blindly Unlearning Backdoors Without Clean Data

Wenbin Jiang, Zihao Wang, Jiqiang Liu, Rui Zhu et autres

Backdoor attacks pose a serious and emerging threat to deep neural networks. By injecting trigger-labeled samples into the training data, these attacks cause compromised models to misclassify any input containing the trigger. Existing defenses typically rely on either recovering the trigger to …

cn, sg, us (code pays fourni par la source)

Pair-wise quantum gossip protocol-based decentralized quantum federated learning

Rui Zhang, Jian Wang, Nan Jiang, Celia-Ourida ZAIDI et autres

An Adaptive Consistency-Based Detection Scheme for Label-Flipping Attacks in Low-Altitude Economic Networks

J.H. Wang, Tianhao Liu, Tao Zhang, Jian Wang et autres

With the rapid development of intelligent transportation and low-altitude economy, collaborative distillation has emerged as an efficient distributed learning paradigm that enables multi-terminal cooperation without sharing raw data. However, the open participation nature of such systems also exposes them to label-flipping attacks, …

cn (code pays fourni par la source)

MSG: Stealing data from pruned neural networks via malicious sparsity guidance

Jing Shang, Jian Wang, Kailun Wang, Nan Jiang et autres

cn (code pays fourni par la source)

Phonon-scattering-induced quantum linear magnetoresistance up to room temperature

Nannan Tang, Shuai Li, Yanzhao Liu, Yang Jiayi et autres

The realization of quantum transport effects at elevated temperatures has long intrigued researchers due to the implications for unveiling novel physics and developing quantum devices. In this work, we report remarkable quantum linear magnetoresistance (LMR) in the Weyl semiconductor tellurium at high …

Generative AI-Driven Cross-Layer Covert Communication: Fundamentals, Framework, and Case Study

Tianhao Liu, Jiqiang Liu, Tao Zhang, Jian Wang et autres

Ensuring end-to-end cross-layer communication security in military networks by selecting covert schemes between nodes is a key solution for military communication security. With the development of communication technology, covert communication has expanded from the physical layer to the network and application layers, …

cn, sg, us (code pays fourni par la source)

Protect NTN-IoT Security by Malicious Traffic Detection: A Multidimensional Hypergraph Learning Approach

Xuzeng Li, Tao Zhang, Jian Wang, Zhen Han et autres

The vast number of devices and the complexity of requirements present significant challenges in ensuring the security of Non-Terrestrial Internet of Things (NT-IoT). Although existing studies have proposed methods like to defend against data theft and network interference attacks, there is still …

cn, sg (code pays fourni par la source)

Zero on-site testing on GNSS for train control towards digital railways

Debiao Lu, Wensheng Qiao, Baigen Cai, Jian Wang et autres

The integration of Global Navigation Satellite System (GNSS) technology into railway train control systems is a crucial step toward achieving the vision of a digital railway. Traditional train control systems undergo extensive in-house tests and prolonged field tests for certification and approval …

cn, us (code pays fourni par la source)